Struct Qdrant 

pub struct Qdrant { /* private fields */ }

Available on crate feature qdrant only.

A struct representing a Qdrant client with configuration options.

This struct is used to interact with the Qdrant vector database, providing methods to create and manage vector collections, store data, and ensure proper indexing for efficient searches.

Can be cloned with relative low cost as the client is shared.

Implementations

impl Qdrant

pub fn builder() -> QdrantBuilder

Returns a new QdrantBuilder for constructing a Qdrant instance.

pub fn try_from_url(url: impl AsRef<str>) -> Result<QdrantBuilder>

Tries to create a QdrantBuilder from a given URL. Will use the api key in QDRANT_API_KEY if present.

Returns

Arguments

• url - A string slice that holds the URL for the Qdrant client.

Returns

A Result containing the QdrantBuilder if successful, or an error otherwise.

Errors

Errors if client fails build

pub async fn create_index_if_not_exists(&self) -> Result<()>

Creates an index in the Qdrant collection if it does not already exist.

This method checks if the specified collection exists in Qdrant. If it does not exist, it creates a new collection with the specified vector size and cosine distance metric.

Returns

A Result indicating success or failure.

Errors

Errors if client fails build

pub fn client(&self) -> &Arc<Qdrant>

Returns the inner client for custom operations

Trait Implementations

impl Clone for Qdrant

fn clone(&self) -> Qdrant

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Qdrant

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Persist for Qdrant

fn batch_size(&self) -> Option<usize>

Returns the batch size for the Qdrant storage.

Returns

An Option<usize> representing the batch size if set, otherwise None.

fn setup<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Sets up the Qdrant storage by creating the necessary index if it does not exist.

Returns

A Result<()> which is Ok if the setup is successful, otherwise an error.

Errors

This function will return an error if the index creation fails.

fn store<'life0, 'async_trait>( &'life0 self, node: TextNode, ) -> Pin<Box<dyn Future<Output = Result<TextNode>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Stores a single indexing node in the Qdrant storage.

WARN: If running debug builds, the store is blocking and will impact performance

Parameters

• node: The TextNode to be stored.

Returns

A Result<()> which is Ok if the storage is successful, otherwise an error.

Errors

This function will return an error if the node conversion or storage operation fails.

fn batch_store<'life0, 'async_trait>( &'life0 self, nodes: Vec<TextNode>, ) -> Pin<Box<dyn Future<Output = IndexingStream<String>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Stores a batch of indexing nodes in the Qdrant storage.

Parameters

• nodes: A vector of TextNode to be stored.

Returns

A Result<()> which is Ok if the storage is successful, otherwise an error.

Errors

This function will return an error if any node conversion or storage operation fails.

type Input = String

type Output = String

fn name(&self) -> &'static str

impl Retrieve<HybridSearch<Filter>> for Qdrant

Implement the Retrieve trait for HybridSearch search strategy.

Can be used in the query pipeline to retrieve documents from Qdrant.

Expects both a dense and sparse embedding to be set on the query.

fn retrieve<'life0, 'life1, 'async_trait>( &'life0 self, search_strategy: &'life1 HybridSearch<Filter>, query: Query<Pending>, ) -> Pin<Box<dyn Future<Output = Result<Query<Retrieved>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn name(&self) -> &'static str

impl Retrieve<SimilaritySingleEmbedding<Filter>> for Qdrant

Implement the Retrieve trait for SimilaritySingleEmbedding search strategy.

Can be used in the query pipeline to retrieve documents from Qdrant.

Supports filters via the qdrant_client::qdrant::Filter type.

fn retrieve<'life0, 'life1, 'async_trait>( &'life0 self, search_strategy: &'life1 SimilaritySingleEmbedding<Filter>, query: Query<Pending>, ) -> Pin<Box<dyn Future<Output = Result<Query<Retrieved>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn name(&self) -> &'static str

impl Retrieve<SimilaritySingleEmbedding> for Qdrant

Ensures that the SimilaritySingleEmbedding search strategy can be used when no filter is set.

fn retrieve<'life0, 'life1, 'async_trait>( &'life0 self, search_strategy: &'life1 SimilaritySingleEmbedding, query: Query<Pending>, ) -> Pin<Box<dyn Future<Output = Result<Query<Retrieved>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn name(&self) -> &'static str